Tentative breakpoints and areas of technical uncertainty for early reading automated disc diffusion for Enterobacterales.

Background: Disc diffusion is a reliable, accurate and cost-efficient procedure for antimicrobial susceptibility testing (AST) but requires long (18-24 h) incubation times. Reading of disc diffusion after short incubation times (6-8 h) by automated systems is feasible but should be categorized with time-adapted breakpoints to reduce errors. Objectives: This study systematically compared early readings (6 and 8 h) of disc diffusion using an automated system with that of the standard 18 h EUCAST method. Time-adapted tentative breakpoints were proposed to discriminate susceptible from resistant isolates and areas of technical uncertainty were defined to minimize the risk of errors. Methods: A total of 1106 Enterobacterales isolates with a wide variety of resistance mechanisms and resistance profiles were included. All isolates were analysed for susceptibility to amoxicillin/clavulanic acid, ceftriaxone, cefepime, meropenem, ciprofloxacin and gentamicin using the automated WASPLab (TM) system. Part of the collection (515 isolates) was also analysed for susceptibility to an additional 10 antibiotics. Results: Separation between WT and non-WT populations was poorer at early incubation times than following standard incubation. Editing of rapid automated AST results after 6 and 8 h incubation with time-adapted breakpoints resulted in 84.0% and 88.5% interpretable results with assignment to the resistant or susceptible category. Major error and very major error rates for the 6 h readings were only 0.4% and 0.3%, virtually identical to those of 18 h AST reading. Conclusions: Time-adapted clinical breakpoints in disc diffusion testing for Enterobacterales allow for accurate automated AST interpretation after shortened incubation times for a large number of antibiotics, with the additional possibility of subsequent confirmation after 18 h incubation.